The Panama Papers reveals a complex network of money laundering whose power lies in its fragmentation: Even if a small piece of it is exposed and “burned”, the rest of the system continues to operate undisturbed

In 2016, the world‘s biggest data leak scandal, known as “The Panama Papers”, surfaced a vast global network – involving celebrities and state leaders – which used offshore tax havens, anonymous transactions and intermediaries and ghosts companies to hide wealth and avoid taxes for billionaires.

Researchers at the USC Viterbi School of Engineering conducted a detailed study and analysis of companies and their relationships, revealing 11.5 million files leaked to the International Consortium of Investigative Journalists (ICIJ).

Academic researchers have made a number of discoveries about the operation of this network and transactions: revealing a uniquely fragmented structure – very different from the more traditional social ones:

This fragmented structure is also the strength of these trading networks as they make it difficult to access them.

Lead author Dr Mayank Kejriwal, an assistant professor in the Department of Industrial and Systems Engineering at Daniel J. Epstein University and the USC Information Sciences Institute, analyzes complex systems such as social network analysis.

The goal of the research team was to study the “Panama Papers” network as a whole, with the same approach that one could consider the grid of relationships in a social network such as Facebook. The interactions between the network poles offer conclusions about how money is being trafficked.

“In general, on any social network like LinkedIn or Facebook, there is something called the ‘Small World Phenomenon,'” says Kejriwal. It means every person on the planet can be connected to anyone else through six intermediate contacts: “For example, you can get to Bill Gates, with about six connections,” he said.

However, the team found that the Panama Papers network had almost rejected this traditional network organizational behavior. Instead of a network of integrated connections, researchers have uncovered a number of secret fragmented connections, with intermediaries and individuals involved in transactions and companies having very few connections to others in the system.

“It was really unusual. “The degree of fragmentation of the network is something I have never seen before,” said Kejriwal. “I do not know of any other network that has this kind of fragmentation.”

“So if you try to find the connection between one organization and another, it would be impossible to find it, because chances are there is no connection – they are completely disconnected,” said Kejriwal.

Most social, friendly, or organizational networks contain a series of triangular structures, a system known as the “friend-friend phenomenon.” “The simple idea is that a friend’s friend is also a friend,” Kejriwal added. “And we can count it by counting the number of triangles in the grid.”

However, the team discovered that this triangular structure was not typical of the Panama Papers network.

“It turns out that not only is it not widespread, but it is much less prevalent, even for a random network,” Kejriwal further explained. “If you connect literally random things, randomly and then count the triangles on this grid, that grid is even harder to access. “Compared to a random network, in this type of network, the links between offshore companies are shuffled until they can be virtually undetectable (so that anyone can trade with anyone else).”

This type of disconnection that makes the system of secret global financial transactions so powerful. The network could not be easily breached, as there is no way to detect relationships between companies.

Dr Kejriwal believes the network’s behavior demonstrates that those involved in the Panama Papers’ network of offshore companies and illegal transactions were smart enough to know exactly how to transfer money in a way that was undetectable and vulnerable through their connections to other systems. Because it is a global network, there are few options for national or international organizations to intervene to combat corruption and money laundering.

* The research has been published in Applied Network Science.